Synapse formation and maturation require signaling between the synaptic partners and signal transduction within each of them. To learn how assembly of the presynaptic neurotransmitter release apparatus is assembled, we focused on two genes, SAD-A and SAD-B. They are the mammalian orthologues of SAD-1, a gene required for presynaptic differentiation in C. elegans. Because SADs are kinases, we hope they will provide a valuable starting point for elucidating regulatory mechanisms that govern assembly of nerve terminals. Unfortunately, initial genetic tests of this idea gave complex results because the two genes play redundant roles and are involved in multiple steps in neuronal development and because SAD-A/B double mutants die at birth, before most synapses have formed. We therefore developed two genetic strategies to circumvent these limitations of pleitropy and lethality. First, we generated a conditional allele to ablate expression in selected neuronal types. Second, we generated alleles sensitive to a specific inhibitor, which allows precise temporal control of SAD-A/B activity and facilitates substrate identification. Using these new reagents, we have obtained preliminary results indicating that SADs are indeed required for complete presynaptic differentiation of several and perhaps most synaptic types. Here we propose to confirm and extend these results, and to initiate tests of our hypothesis that SAD kinases are critical components of pathways that lead from target-derived synaptic organizing molecules to assembly of nerve terminals. First, we will use the conditional allele to bypass neonatal lethality and characterize presynaptic defects in four peripheral and central excitatory synaptic types. We will also ask whether SADs are also required for development of inhibitory synapses, and whether SADs regulate post- as well as presynaptic development. Second, we will assay synaptic development and function with SAD-A and -B mutant alleles that render the kinases selectively inhibitable by an ATP analog to which unmodified kinases are insensivitive. These alleles provide us with precise and reversible temporal control over SAD kinase activity, both in vivo and in cultures generated from the mutant mice. We can therefore ask when during development SADs are required, whether their activity is required for synaptic maintenance in adults, and whether acute inhibition of the kinases affects the function of synapses that have developed normally. Finally, we will initiate studies aimed at learning how synaptogenic signals activate SADs and how SADs, in turn, coordinate presynaptic differentiation.